A file abstraction for sequential writing. The implementation must provide buffering since callers may append small fragments at a time to the

return tensorflow::Status::OK()

tensorflow::RandomAccessFile::RandomAccessFile()

tensorflow::TensorShape::TensorShape(gtl::ArraySlice< int64 > dim_sizes) Construct a TensorShape from the provided

size_t tensorflow::ThreadOptions::stack_size Thread stack size to use (in bytes).

void tensorflow::Status::Update(const Status &new_status) If ok(), stores new_status into *this

TTypes< T, NDIMS >::Tensor tensorflow::Tensor::flat_inner_dims() Returns the data as an Eigen::Tensor with NDIMS dimensions

Member Details tensorflow::Status::Status() Create a success status.

TTypes<T>::ConstVec tensorflow::Tensor::vec() const Const versions of all the methods above.

Status tensorflow::Env::IsDirectory(const string &fname) Returns whether the given path is a directory or not. Typical return